Models predicting effects of climate change on oceans take top prizes at Compute Canada competition
From left to right: Brian Stewart, Laura Castro de la Guardia and Dan St. Germain
Two University of Alberta-led research projects that used mathematical simulations to provide a critical window into the complex ecosystem changes happening in our northern oceans have won the top two prizes at the CANHEIT | HPCS Conference in Edmonton on June 22.
Organized by Compute Canada, this year’s research poster competition received over 20 submissions from scientists from a variety of scientific disciplines, from pharmaceutical sciences to geomatics engineering. A total of four prizes were awarded based on their visual design, science, relevance, how well they identified and addressed the problem, research methodology, and the overall presentation of the results. The top three winners were selected by a panel of three judges and a fourth prize, the People’s Choice Award, was decided in online voting by CANHEIT | HPCS delegates.
“The judges were impressed with the high calibre of projects that presented this year. Picking just a few winners wasn’t easy,” says Dr. Dugan O’Neil, Chief Science Officer at Compute Canada. “The ingenuity, skill and passion of our young scientists are resulting in world-leading research that will help Canada and other countries take steps to improve human health, the environment and other grand challenges.”
This year’s top prize went to a University of Alberta team led by Ph.D. student Laura Castro de la Guardia, who came to Canada from Cuba to study Arctic marine ecosystems at the intersection of biology and oceanography. The research team’s enhanced numerical models produced high-resolution simulations of historical physical and biogeochemical ocean processes (1958-2013) that can be applied to the entire North Atlantic, particularly the Labrador Sea. The models feed into a growing evidence base that will help scientists and policymakers better predict the effects of global warming on marine ecosystems.
Castro was also a contributing researcher on the project that won second place. Led by Dr. Xianmin Hu, who came to Alberta from China in 2007, the team demonstrated two numerical ocean models developed with computational support from WestGrid and Compute Canada. The project’s results provide valuable knowledge on the impact of Greenland’s melting ice sheet and future ecosystem changes under different climate scenarios. The group also examined the computation costs of these major computer simulations.
The third place prize was awarded to a University of Alberta team who showed how a computer program they designed to play a nearly perfect game of heads-up limit Texas hold’em poker can be adapted for more complex and serious “games”, namely optimizing security deployments to prevent a terrorist or cybersecurity attack. The mathematical algorithm, developed by Finnish software developer Oskari Tammelin, produced results faster than current artificial intelligence techniques, while using five times less computation time. Those advances would have never occurred had it not been for the scale of resources provided by Compute Canada and Calcul Québec: 1000 years of processing power, 200 distributed high performance computers and 17 terabytes of data storage.
This year’s People’s Choice Award goes to a University of Alberta-led team that developed a virtual model of the heart’s cellular channels that could help medical researchers study new drug therapies using computational approaches. Lead author Horia Jalily Hasani, who hails from Iran, worked with her colleagues to develop one of the most detailed computer models of a potassium ion channel. The new tool allows researchers to study mutations in a gene responsible for making potassium channels, which play a role in recharging the cardiac muscle after each heartbeat to maintain a regular rhythm. Mutations in the KCNQ1 gene have been implicated in a wide range of cardiac diseases, including Long QT syndrome, a mutation in the heart’s electrical system that can cause the heart to beat irregularly.
This year’s competition was unique due to the opening of the submissions to research involving ARC as well as that related to Higher Education IT solutions
The following provides hotlink to the full research abstracts and a full list of authors.
NEMO3.4 — BLINGv0, an ocean-sea ice-biogeochemical model for the study of gasses and ocean productivity: Laura Castro de la Guardia (University of Alberta), Xianmin Hu (University of Alberta), Mariona Claret (University of McGill), Nathan Grivault (University of Alberta), Paul G. Myers (supervisor, University of Alberta), Eric D. Galbraith (supervisor: Institution Catalana de Recerca i Estudis Acancats (ICREA)), Yarisbel Garcia (University of Alberta), Laura C. Gillard (University of Alberta), Natasha Ridenour (University of Alberta), Clark Pennelly (University of Alberta), Juliana Marson (University of Alberta)
Ocean General Circulation Modelling Using NEMO and the ANHA Configuration: Xianmin Hu (University of Alberta), Clark Pennelly (University of Alberta), Juliana Marson (University of Alberta), Laura Castro De La Guardia(University of Alberta), Laura C. Gillard (University of Alberta), Natasha Ridenour (University of Alberta), Nathan Grivault (University of Alberta), Samantha Roch (University of Alberta), Yarisbel Garcia (University of Alberta), Paul G. Myers (University of Alberta)
Modeling the Human KCNQ1 Potassium Ion Channel: Application of Computational Approaches to Cardiotoxicity: Horia Jalily Hasani (Faculty of Pharmacy and Pharmaceutical Sciences, University of Alberta, Edmonton, AB, Canada.), Marawan Ahmed (Faculty of Pharmacy and Pharmaceutical Sciences, University of Alberta, Edmonton, AB, Canada.), Khaled Barakat (*Faculty of Pharmacy and Pharmaceutical Sciences, University of Alberta, Edmonton, AB, Canada. *Li Ka Shing Institute of Virology, University of Alberta, Edmonton, AB, Canada)